New opportunity

AI Manufacturing Excellence Intern

BD · Jurong East, Singapore

About this role

BD is seeking an AI Manufacturing Excellence Intern to design and develop AI assistants for manufacturing productivity. This role involves building solutions on approved enterprise platforms, developing retrieval logic, and creating safeguards for AI outputs. Responsibilities include: AI Solution Development: Design and develop AI assistants using platforms like Microsoft Copilot Studio or Azure AI Foundry, ensuring solutions are built on certified data layers with traceable sources. Opportunity Identification: Analyze productivity and cost measures to identify performance gaps and improvement areas, providing evidence-based recommendations. Knowledge Management: Create and maintain a reusable prompt library for productivity analysis and root-cause exploration. Plant Adoption: Conduct AI maturity assessments, run pilot sessions, and provide training for plant engineers to ensure successful adoption. Required Qualifications: Currently pursuing a degree in Engineering, Industrial Engineering, Data Science, Computer Science, Information Technology, Business Analytics, or a related field. Must be available for a full-time six-month internship. Practical exposure to Python, Power Platform, and basic SQL is required. Candidates must be willing to travel to manufacturing sites at Tuas Avenue 2 when necessary, with the primary work location at Jurong East. Preferred Qualifications: Experience with Microsoft Copilot Studio, Azure AI Foundry, or similar agent platforms. Understanding of RAG and generative AI failure modes. Experience with Lean Six Sigma, Kaizen, or manufacturing projects, as well as facilitation of user testing and training. Proficiency in Power BI, SharePoint, and Teams is a plus.

For full details about this role, click “Apply on company site” to view the employer’s original posting.

Skills for this role

PythonPower PlatformSQLMicrosoft Copilot StudioAzure AI FoundryRAGGenerative AILean Six SigmaKaizenPower BISharePointMicrosoft 365

Checking your profile…

Preparation suggestions, not the employer’s actual interview questions. Always check the original posting for current requirements.

Understand the role

What matters most

AI Solution Development and Governance
The primary focus is designing AI assistants using platforms like Microsoft Copilot Studio or Azure AI Foundry. You must ensure all solutions are built on certified data layers, incorporate retrieval logic with traceable sources, and implement strict safeguards to prevent unsupported responses or data leakage.
Manufacturing Productivity Analysis
You are expected to analyze productivity and cost measures—such as downtime, scrap, and labor hours—to identify performance gaps. This requires connecting operational losses to cost drivers and building recommendation capabilities that retrieve relevant historical improvement cases like A3 or Kaizen.
Plant Adoption and Capability Building
Success is measured by how well plant engineers adopt these AI tools. You must conduct maturity assessments, facilitate pilot sessions, and create training materials to ensure AI solutions integrate into existing workflows rather than creating parallel, disconnected processes.

Likely challenges to think through

Balancing AI Innovation with Data Integrity
A significant challenge is the requirement to avoid direct connections to raw transactional systems while still providing useful, evidence-based insights. Prepare by demonstrating how you manage data abstraction layers and ensure that AI outputs remain grounded in certified, reliable data sources.
Bridging Technical AI and Operational Excellence
You must translate complex AI retrieval logic into practical, actionable advice for plant engineers who may be unfamiliar with generative AI. Prepare by practicing how to explain technical concepts like RAG and prompt engineering in the context of Lean Six Sigma or Kaizen methodologies.

Practice your answers

Technical practice

RAG Architecture and Failure Modes

What to coverExplain how Retrieval-Augmented Generation (RAG) works in an enterprise context. Discuss common failure modes such as hallucinations, retrieval of irrelevant documents, or lack of source traceability, and how you would mitigate these in a manufacturing assistant.

Data Governance and Security in AI

What to coverThe role requires building on certified data layers without connecting to raw transactional systems. Describe the importance of data governance and how you would ensure an AI assistant respects access restrictions and protects confidential manufacturing data.

Prompt Engineering for Manufacturing

What to coverHow would you design a prompt library to ensure consistency across different plant teams? Discuss the components of a robust prompt, such as defining inputs, expected output formats, and constraints to prevent unsupported responses.

SQL and Data Analysis for Productivity

What to coverGiven a dataset of manufacturing downtime and scrap rates, how would you use SQL to identify performance gaps? Explain how you would translate these technical findings into actionable insights for plant engineers.

AI Maturity Assessments

What to coverIf tasked with conducting an AI maturity assessment at a manufacturing site, what key indicators would you look for? Suggest factors such as data availability, existing digital tools, and the team's readiness to adopt new AI-driven workflows.

Integrating AI with Lean Six Sigma

What to coverHow can AI assistants support methodologies like DMAIC or Kaizen? Explain how you would structure a retrieval system to surface relevant historical improvement cases based on specific waste categories or process types.

Testing and Validation Strategies

What to coverDescribe how you would build a structured test set for an AI assistant. Include how you would handle edge cases and ensure the model provides citations for quantitative outputs to maintain trust with plant operators.

Power Platform for Operational Excellence

What to coverHow would you leverage the Power Platform to support the adoption of AI solutions? Suggest ways to integrate AI assistants into existing workflows using tools like Power BI for visualization or Teams for user interaction.

Behavioural practice

Managing User Resistance to AI

What to coverDescribe a time you had to introduce a new tool or process to a group that was accustomed to traditional methods. How did you handle skepticism, and what steps did you take to ensure successful adoption?

Translating Technical Concepts to Non-Technical Users

What to coverProvide an example of when you had to explain a complex technical solution to a stakeholder without a technical background. How did you ensure they understood the value and limitations of your work?

Prioritizing Tasks in a Dynamic Environment

What to coverIn an internship, you may balance multiple pilot sessions and development tasks. Describe how you prioritize your work when faced with competing demands from different plant engineers or project leads.

Learning from Failure in Projects

What to coverTell me about a project where the initial approach did not yield the expected results. How did you analyze the failure, what did you learn, and how did you adjust your strategy to move forward?

Sharpen your knowledge

Topics to prepare

Retrieval-Augmented Generation (RAG) Architecture
Understand how RAG systems retrieve external data to ground LLM responses. Exercise: Design a conceptual workflow for an AI assistant that retrieves manufacturing productivity data from a secure database, ensuring citations are included in the output. This is critical for the role's requirement to build assistants with traceable sources.
Generative AI Failure Modes and Safeguards
Study common issues like hallucinations, data leakage, and prompt injection. Exercise: Create a list of potential failure modes for a manufacturing assistant and draft specific system-level safeguards or prompt constraints to mitigate them. This aligns with the responsibility to create safeguards for AI outputs.
Lean Six Sigma and Continuous Improvement (CI) Methodologies
Review core concepts like DMAIC, Kaizen, and A3 problem-solving. Exercise: Map how an AI assistant could support a DMAIC project phase, such as identifying cost drivers like downtime or scrap. This is relevant to the role's focus on analyzing productivity and recommending improvements.

Skills to revise

Microsoft Copilot Studio and Azure AI Foundry
Gain familiarity with building agents on enterprise platforms. Exercise: Explore the documentation for creating custom agents in Copilot Studio or Azure AI Foundry, focusing on how to connect data sources and configure retrieval logic. This is a preferred qualification for the role.
SQL for Data Analysis
Practice querying structured data to identify performance gaps. Exercise: Write SQL queries to calculate key manufacturing metrics like labor hours per unit or scrap rates from a sample dataset. This is essential for the requirement to analyze productivity and cost measures.
Prompt Engineering for Knowledge Management
Develop structured, reusable prompts for specific workflows. Exercise: Create a library of prompts designed to extract root-cause analysis from manufacturing reports, ensuring consistent output formats. This directly supports the task of maintaining a reusable prompt library.

Your preparation checklist

Review RAG and Generative AI Fundamentals
Prepare to discuss how Retrieval-Augmented Generation (RAG) works in an enterprise context. Focus on understanding common failure modes like hallucinations, data leakage, and the importance of traceable citations when using LLMs for technical manufacturing data.
Prepare a Manufacturing Process Improvement Case Study
Draft a brief walkthrough of a project where you applied Lean Six Sigma or Kaizen principles. Highlight how you identified a performance gap (e.g., downtime or scrap) and used data to recommend a solution, as this mirrors the intern's responsibility to connect operational losses to cost drivers.
Demonstrate Technical Proficiency in Python and SQL
Prepare code samples or a portfolio project that demonstrates your ability to clean and query datasets using SQL and perform data analysis or automation tasks using Python. Be ready to explain how you would structure data to be consumed by an AI assistant.
Familiarize with Microsoft Power Platform and AI Tools
Review documentation for Microsoft Copilot Studio and Azure AI Foundry. If possible, create a simple prototype or workflow in Power Platform to show your familiarity with low-code enterprise tools and how they integrate with data sources like SharePoint or Power BI.
Develop Strategic Questions for the Interviewer
Prepare thoughtful questions regarding the team's approach to AI governance, such as: 'How does the team balance the need for rapid AI adoption with the strict requirement for certified data layers?' or 'What are the most common barriers to AI maturity you have observed in the plant environments?'

Availability has not yet been verified. Status can change after a check; confirm on the employer’s website. Report an issue · Our checking policy

KEEP EXPLORING

Similar AI jobs

Related skills and specializations in Singapore. Matched to this role, not your profile.

Explore more jobs

Finding similar opportunities…